: In recent years, there has been increased recognition of adolescent major depressive disorder (MDD) as a severe, highly prevalent and chronically disabling disorder. By most measures, adolescent MDD is continuous with adult MDD, including clinical course, secular trends, and psychobiologic correlates. Studies of the illness during adolescence can, therefore, clarify the contribution of neurodevelopmental abnormalities to the pathogenesis of the disorder. Neurobiological models for MDD have consistently implicated temporo-limbic, prefrontal cortical and basal ganglia circuits in the pathophysiology of the disorder. Strikingly, children and adolescents with MDD have been largely overlooked in brain imaging studies. In particular, there are no published studies using magnetic resonance spectroscopy (MRS) to investigate brain chemistry in depressed children and adolescents. Yet the inclusion of children and adolescents in these types of studies is critical in order to determine whether abnormalities in brain chemistry in depressed patients are the consequence of altered development over the life span, or if they are present even early in development thus representing a risk factor for subsequent development of depression. Preliminary studies by the applicants using MRS suggest increased caudate and anterior cingulate choline concentrations in psychotropic-naive pediatric MDD patients compared to age and sex matched controls. No differences in choline levels were observed between MDD patients and controls in occipital cortex choline levels. Pilot studies showed reductions in caudate but not occipital choline concentrations after 12 weeks of effective monotherapy with the selective serotonin reuptake inhibitor (SSRI) paroxetine. In this project, proton MRS will be used for the direct, in vivo and noninvasive evaluation and comparison of the impact of paroxetine vs. cognitive behavioral therapy (CBT) on brain chemistry as related to treatment efficacy in adolescents with MDD. Such an approach will help 1) determine whether psychotropic-naive adolescents with MDD differ from healthy controls in neurochemistry and 2) determine whether effective treatment with paroxetine and CBT results in localized and treatment-specific changes in brain chemistry. An understanding of the neurochemical disturbances in MDD may be important for the development of new treatment approaches.